Accurate detection and anatomic localization of both primary and metastatic lesions remains one of the major problems in the management of most human carcinomas. We have recently initiated clinical trials at the NIH Clinical Center to detect and localize colorectal carcinoma lesions using radiolabeled monoclonal antibody (MAb) B72.3. Parameters that are being systematically investigated concerning both the efficiency of MAb localization and the efficiency of gamma scanning of carcinoma lesions include: (a) effect of MAb dose and specific activity of radionuclide coupled MAb; (b) comparison of the use of intact IgG, F(ab')2, and Fab'; (c) choice of radionuclide; (d) route of inoculation; (e) size, location, and other inherent properties of the tumor mass such as antigen content; (f) the presence of circulating antigen; (g) the presence and/or absence of human anti-murine Ig antibodies; (h) metabolism of MAb and fragments; (i) combinations of MAbs. It is hoped that these studies will also aid in establishing a rational basis for the subsequent therapeutic use of a particular MAb, either coupled to toxins, via effector cell-mediated or complement-mediated mechanisms, or using MAbs radiolabeled with one of a variety of isotopes. This latter goal can be accomplished by direct analyses of biopsy material (both tumor and normal tissues) from patients receiving radiolabeled MAb to define the "radiolocalization index" or potential "therapeutic index" (i.e., the ratio of the amount of MAb bound (via cpm) per gram of tumor tissue to that bound per gram of normal tissues).